478 Mr. G. A. K. Marshall on 



for supposing that the lighter Asiatic form is the older 

 [see pp. 4-71, 470]. On general principles it would seem 

 that swamping would be likely to keep the species more 

 or less constant in its ancestral home, whereas those speci- 

 mens that wandered further afield would probably tend 

 to vary along slightly different lines ; but perhaps I have 

 not ])roperly caught your idea. The case of ahippus 

 would be a great deal more difficult to explain satisfactor- 

 ily, seeing that it occurs also at Aden ; and Butler says 

 that examples sent from such places as Monbuttu, Wade- 

 lai, etc., by Emin Pasha, showed every gradation from 

 chrysippus through alcippfAdcs to alcippus ; further, if I 

 remember rightly, you wrote me that an example I sent 

 you from the Tugela had the white developed as strongly 

 as in any West Coast s^^ecimen, and Burn said they were 

 by no means uncommon there." 



["A. G. Butler records (Proc. Zool. Soc, 1806, p. 243) 

 Captain Nurse's statement that in Somaliland he bred 

 all four forms of clirysippits from quite similar larvie." 

 G. A. K. M., 1002.] 



" Salishiirij, June 20, 10(10. — Referring to the cpiestion 

 of Limnas IduyU, although I fully appreciate the value of 

 your arguments, yet I nmst confess that when looking at 

 the matter from the point of view of an opponent of 

 mimicry, it seems at least open to criticism. The difficulty 

 seems to lie in the fact that the same coloration would 

 thus have to be regarded as both protective and warning. 

 Now you have said that in desert regions insects would be 

 more liable to attack owing to the paucity of insect life, 

 and I should be glad to know whether you have any 

 special reasons for adopting this view, as I have no ex- 

 perience of what the conditions of life really are in such 

 localities. But don't you think that it is more likely that 

 the struggle for existence would be principally against 

 climatic conditions and not so much a competition with 

 other organisms, and that thus probably insects would 

 have a hQiiQV p)roportionate chance of finding a living than 

 would the vertebrates as compared with more fertile 

 regions ? If this were so it would follow that insects 

 would be comparatively freer from attack in desert regions, 

 and this would atford us another explanation of the Idugii 

 phenomenon. We might presume that the less conspicuous 

 hlugii colouring was the more ancestral (as seems not 

 unlikely), but that in the more fertile regions where 



